Combining the results of several neural network classifiers
Neural Networks
Machine Learning
Combining labeled and unlabeled data with co-training
COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory
The Random Subspace Method for Constructing Decision Forests
IEEE Transactions on Pattern Analysis and Machine Intelligence
Machine Learning
Hierarchical neural networks utilising dempster-shafer evidence theory
ANNPR'06 Proceedings of the Second international conference on Artificial Neural Networks in Pattern Recognition
| Hi-index | 0.00 |
In this paper, a method to construct ensembles of tree-structured classifiers using multi-view learning instead of feature subset selection is proposed. An essential requirement to train an effective classifier ensemble is the diversity among the base individual classifiers. In order to construct diverse individual classifiers, it is assumed that the object to be classified is represented through multiple feature sets (views). Different views of the data can then be combined to improve the accuracy of the learning task. Multi-view Forests have been examined on a benchmark data set of 3D-object recognition based on 2D camera images. Results show that multi-view learning can improve the performance of the individual classifiers and the whole ensemble, and thus promoting the use of multi-view forests.